license: apache-2.0
datasets:
- DevQuasar/brainstorm_vicuna_1k
base_model: meta-llama/Meta-Llama-3-8B
pipeline_tag: text-generation
model_context_window: 8192
Model intention
Enables user to talk through an idea with an LLM model, where the model is asking various questions about the idea that helps users to drill down into the details and/or rethink the idea.
The model is not intended to provide answers for quuestions, but the other way around: Ask question from the user that force the user to destil the idea.
Examples
Inference code
https://github.com/csabakecskemeti/ai_utils/blob/main/brainstorm_inference.py
Example generations
[USER] (Ctrl+C to exit): Today I'm gonna buy all healthy product at the grocery store.
[CHAT]: What kind of healthy products are you thinking of buying?
[USER] (Ctrl+C to exit): Mainly vegetables, and some lean white meet like fish and chicken
[CHAT]: Have you considered organic produce for your veggies?
[USER] (Ctrl+C to exit): I'm thinking to invest in a franchise to get a new source of income.
[CHAT]: That's an interesting idea! What kind of franchises are you considering?
[USER] (Ctrl+C to exit): Something that is a small upfront cost, not sure what would be a great choice.
[CHAT]: Have you considered food-related franchises like pizza or ice cream shops? They often have low startup costs.
[USER] (Ctrl+C to exit): But that risky due to publich healt requirements
[CHAT]: You're right, there are some regulations to consider. How would you plan to mitigate those risks?
[USER] (Ctrl+C to exit): I guess do some other non food related franchise
[CHAT]: What about a business service franchise like janitorial or landscaping services? Those might have lower regulatory burdens.